This study analyzed 87 athletes attending the Para Olympic Games at Stoke Mandeville in 1984. All athletes had had complete neurologic lesions between the levels of T10 adn L2. Clinical details were obtained from each athlete. Assessment then consisted of measurement of spinal movement in the sagittal plane using spondylometers and movement in the rotation plane using a rotameter. Body trunk strength was measured with the athletes in their own wheelchairs using a myometer. This gave a recording of kilograms of force. Trunk balance was assessed on both a static and a dynamic basis. The age, sex, follow-up period, and body weight for both spinal fusion and conservatively treated groups were similar. There was little difference in the incidence of pain between those treated by conservative and operative methods. There was statistically significant difference in the range of sagittal plane and rotation movement. There was no difference in flexor trunk power measured with the myometer between the two groups, but when extensor power was measured it was found to be 25% less in the spinal fusion group. There was no difference between the two groups when tested for static and dynamic trunk balance. Two other athletes who had had spinal fusions extending from the upper thoracic region to the sacrum showed unusually poor trunk stability. Overall, this study demonstrates that spinal fusion, particularly over multiple segments in complete paraplegics, has a deleterious effect not only on spinal movement but also on body trunk strength.